Bulimia nervosa is a prevalent form of psychopathology among late adolescent and young adult women. The disorder is associated with significant adverse medical and psychosocial sequelae. Although manual based psychotherapies for bulimia nervosa have been developed and a standard-of-care for these patients has emerged, most practicing psychotherapists who treated individuals with bulimia nervosa have been developed and a standard-of-care for these patients has emerged most practicing psychotherapists who treat individuals with bulimia nervosa have not been adequately trained to deliver such therapies. Therefore, there appears to be a growing discrepancy between that is being used experimentally in academic centers and recommended by researchers in the field (e.g., cognitive behavioral psychotherapy), and what is actually available in the community. Although this deficiency might be remedied through the development of training programs to improve therapists skills in these areas, such programs would be difficult to administer logistically, and it is unlikely that most practitioners would have the time or financial resources to undertake such training. Also, the skills obtained might trophy over time if a sufficient number of cases were not seen later, a likely outcome for many therapists practicing in smaller urban and rural areas. Other models to deliver such therapy would be to have trained therapists travel to these area on a regular basis, although given current reimbursement guidelines such an approach would probably not be self-supporting, and given the amount of travel time that would be necessary, most psychotherapists would not wit to fill such a position on an ongoing basis. A model that would seem to incorporate the best of the "on-site" delivered CBT, yet not require the therapist to travel (which in areas such as North Dakota and Northwestern Minnesota can involve substantial distances) is to deliver cognitive behavioral therapy using AV technology. In this application we propose to test the effectiveness and acceptability of the on- site CBT model versus CBT delivered through AV technology. The implications of this study could be very important for delivery of mental health services in non-urban areas and in states with low population densities. If effectiveness can be demonstrated and the method of delivery is acceptable to patients it would be possible to develop a menu of various mental health services that could be provide through V technology to rural areas.